Water and Plant Cells3Chapter

WATER PLAYS A CRUCIAL ROLE in the life of the plant. For everygram of organic matter made by the plant, approximately 500 g of wateris absorbed by the roots, transported through the plant body and lost tothe atmosphere. Even slight imbalances in this flow of water can causewater deficits and severe malfunctioning of many cellular processes.Thus, every plant must delicately balance its uptake and loss of water.This balancing is a serious challenge for land plants. To carry on photo-synthesis, they need to draw carbon dioxide from the atmosphere, butdoing so exposes them to water loss and the threat of dehydration.

A major difference between plant and animal cells that affects virtuallyall aspects of their relation with water is the existence in plants of the cellwall. Cell walls allow plant cells to build up large internal hydrostaticpressures, called turgor pressure, which are a result of their normal waterbalance. Turgor pressure is essential for many physiological processes,including cell enlargement, gas exchange in the leaves, transport in thephloem, and various transport processes across membranes. Turgor pres-sure also contributes to the rigidity and mechanical stability of nonligni-fied plant tissues. In this chapter we will consider how water moves intoand out of plant cells, emphasizing the molecular properties of water andthe physical forces that influence water movement at the cell level. Butfirst we will describe the major functions of water in plant life.

WATER IN PLANT LIFEWater makes up most of the mass of plant cells, as we can readily appre-ciate if we look at microscopic sections of mature plant cells: Each cellcontains a large water-filled vacuole. In such cells the cytoplasm makesup only 5 to 10% of the cell volume; the remainder is vacuole. Water typ-ically constitutes 80 to 95% of the mass of growing plant tissues. Com-mon vegetables such as carrots and lettuce may contain 85 to 95% water.Wood, which is composed mostly of dead cells, has a lower water con-tent; sapwood, which functions in transport in the xylem, contains 35 to

75% water; and heartwood has a slightly lower water con-tent. Seeds, with a water content of 5 to 15%, are among thedriest of plant tissues, yet before germinating they mustabsorb a considerable amount of water.

Water is the most abundant and arguably the best sol-vent known. As a solvent, it makes up the medium for themovement of molecules within and between cells andgreatly influences the structure of proteins, nucleic acids,polysaccharides, and other cell constituents. Water formsthe environment in which most of the biochemical reac-tions of the cell occur, and it directly participates in manyessential chemical reactions.

Plants continuously absorb and lose water. Most of thewater lost by the plant evaporates from the leaf as the CO2needed for photosynthesis is absorbed from the atmo-sphere. On a warm, dry, sunny day a leaf will exchange upto 100% of its water in a single hour. During the plant’s life-time, water equivalent to 100 times the fresh weight of theplant may be lost through the leaf surfaces. Such water lossis called transpiration.

Transpiration is an important means of dissipating theheat input from sunlight. Heat dissipates because the watermolecules that escape into the atmosphere have higher-than-average energy, which breaks the bonds holding themin the liquid. When these molecules escape, they leavebehind a mass of molecules with lower-than-averageenergy and thus a cooler body of water. For a typical leaf,nearly half of the net heat input from sunlight is dissipatedby transpiration. In addition, the stream of water taken upby the roots is an important means of bringing dissolvedsoil minerals to the root surface for absorption.

Of all the resources that plants need to grow and func-tion, water is the most abundant and at the same time themost limiting for agricultural productivity (Figure 3.1). Thefact that water is limiting is the reason for the practice ofcrop irrigation. Water availability likewise limits the pro-ductivity of natural ecosystems (Figure 3.2). Thus anunderstanding of the uptake and loss of water by plants isvery important.

We will begin our study of water by considering how itsstructure gives rise to some of its unique physical proper-ties. We will then examine the physical basis for watermovement, the concept of water potential, and the appli-cation of this concept to cell–water relations.

THE STRUCTURE AND PROPERTIES OF WATERWater has special properties that enable it to act as a sol-vent and to be readily transported through the body of theplant. These properties derive primarily from the polarstructure of the water molecule. In this section we willexamine how the formation of hydrogen bonds contributesto the properties of water that are necessary for life.

The Polarity of Water Molecules Gives Rise toHydrogen BondsThe water molecule consists of an oxygen atom covalentlybonded to two hydrogen atoms. The two O—H bondsform an angle of 105° (Figure 3.3). Because the oxygenatom is more electronegative than hydrogen, it tends toattract the electrons of the covalent bond. This attractionresults in a partial negative charge at the oxygen end of themolecule and a partial positive charge at each hydrogen.

Chapter 334

10 20 30 40 50 60

2.0

4.0

6.0

8.0

10.0

0

Co

rn y

ield

(m

3 h

a–1)

Water availability (number of days withoptimum water during growing period)

0.5 1.0 1.5 2.0

500

1000

1500

0

Pro

du

ctiv

ity

(dry

g m

–2 y

r–1)

Annual precipitation (m)

FIGURE 3.1 Corn yield as a function of water availability.The data plotted here were gathered at an Iowa farm over a4-year period. Water availability was assessed as the num-ber of days without water stress during a 9-week growingperiod. (Data from Weather and Our Food Supply 1964.)

FIGURE 3.2 Productivity of various ecosystems as a func-tion of annual precipitation. Productivity was estimated asnet aboveground accumulation of organic matter throughgrowth and reproduction. (After Whittaker 1970.)

These partial charges are equal, so the water molecule car-ries no net charge.

This separation of partial charges, together with theshape of the water molecule, makes water a polar molecule,and the opposite partial charges between neighboringwater molecules tend to attract each other. The weak elec-trostatic attraction between water molecules, known as ahydrogen bond, is responsible for many of the unusualphysical properties of water.

Hydrogen bonds can also form between water and othermolecules that contain electronegative atoms (O or N). Inaqueous solutions, hydrogen bonding between water mol-ecules leads to local, ordered clusters of water that, becauseof the continuous thermal agitation of the water molecules,continually form, break up, and re-form (Figure 3.4).

The Polarity of Water Makes It an Excellent SolventWater is an excellent solvent: It dissolves greater amountsof a wider variety of substances than do other related sol-vents. This versatility as a solvent is due in part to the smallsize of the water molecule and in part to its polar nature.The latter makes water a particularly good solvent for ionicsubstances and for molecules such as sugars and proteinsthat contain polar —OH or —NH2 groups.

Hydrogen bonding between water molecules and ions,and between water and polar solutes, in solution effectivelydecreases the electrostatic interaction between the chargedsubstances and thereby increases their solubility. Further-more, the polar ends of water molecules can orient them-selves next to charged or partially charged groups inmacromolecules, forming shells of hydration. Hydrogenbonding between macromolecules and water reduces theinteraction between the macromolecules and helps drawthem into solution.

The Thermal Properties of Water Result fromHydrogen BondingThe extensive hydrogen bonding between water moleculesresults in unusual thermal properties, such as high specificheat and high latent heat of vaporization. Specific heat isthe heat energy required to raise the temperature of a sub-stance by a specific amount.

When the temperature of water is raised, the moleculesvibrate faster and with greater amplitude. To allow for thismotion, energy must be added to the system to break thehydrogen bonds between water molecules. Thus, com-pared with other liquids, water requires a relatively largeenergy input to raise its temperature. This large energyinput requirement is important for plants because it helps

buffer temperature fluctuations.Latent heat of vaporization is

the energy needed to separatemolecules from the liquid phaseand move them into the gas phaseat constant temperature—a processthat occurs during transpiration.For water at 25°C, the heat ofvaporization is 44 kJ mol–1—thehighest value known for any liq-uid. Most of this energy is used tobreak hydrogen bonds betweenwater molecules.

The high latent heat of vapor-ization of water enables plants tocool themselves by evaporatingwater from leaf surfaces, whichare prone to heat up because ofthe radiant input from the sun.Transpiration is an importantcomponent of temperature regu-lation in plants.

Water and Plant Cells 35

H H

O

105°

d–

d+ d+Net positive charge

Attraction of bondingelectrons to the oxygen creates local negativeand positive partial charges

Net negative charge

O

O

O

O

O

O

OO

O

O

OH

H

H

H

HH

H

H H

HH

HH

H

HH

H

HH

H

H

H

H

H

H

H

HH

H

HHH

H

H

H

HH

H

H

H

H

H

O

O

O

O

O

O

O

OO

O

HH

O

(A) Correlated configuration (B) Random configuration

FIGURE 3.3 Diagram of the water molecule. The twointramolecular hydrogen–oxygen bonds form an angle of105°. The opposite partial charges (δ– and δ+) on the watermolecule lead to the formation of intermolecular hydrogenbonds with other water molecules. Oxygen has six elec-trons in the outer orbitals; each hydrogen has one.

FIGURE 3.4 (A) Hydrogen bonding between water molecules results in local aggre-gations of water molecules. (B) Because of the continuous thermal agitation of thewater molecules, these aggregations are very short-lived; they break up rapidly toform much more random configurations.

Chapter 3 36

The Cohesive and Adhesive Properties of WaterAre Due to Hydrogen BondingWater molecules at an air–water interface are more stronglyattracted to neighboring water molecules than to the gasphase in contact with the water surface. As a consequence ofthis unequal attraction, an air–water interface minimizes itssurface area. To increase the area of an air–water interface,hydrogen bonds must be broken, which requires an input ofenergy. The energy required to increase the surface area isknown as surface tension. Surface tension not only influ-ences the shape of the surface but also may create a pressurein the rest of the liquid. As we will see later, surface tensionat the evaporative surfaces of leaves generates the physicalforces that pull water through the plant’s vascular system.

The extensive hydrogen bonding in water also gives riseto the property known as cohesion, the mutual attractionbetween molecules. A related property, called adhesion, isthe attraction of water to a solid phase such as a cell wallor glass surface. Cohesion, adhesion, and surface tensiongive rise to a phenomenon known as capillarity, the move-ment of water along a capillary tube.

In a vertically oriented glass capillary tube, the upwardmovement of water is due to (1) the attraction of water tothe polar surface of the glass tube (adhesion) and (2) thesurface tension of water, which tends to minimize the areaof the air–water interface. Together, adhesion and surfacetension pull on the water molecules, causing them to moveup the tube until the upward force is balanced by theweight of the water column. The smaller the tube, thehigher the capillary rise. For calculations related to capil-lary rise, see Web Topic 3.1.

Water Has a High Tensile StrengthCohesion gives water a high tensile strength, defined asthe maximum force per unit area that a continuous columnof water can withstand before breaking. We do not usuallythink of liquids as having tensile strength; however, such aproperty must exist for a water column to be pulled up acapillary tube.

We can demonstrate the tensile strength of water by plac-ing it in a capped syringe (Figure 3.5). When we push on theplunger, the water is compressed and a positive hydrosta-tic pressure builds up. Pressure is measured in units calledpascals (Pa) or, more conveniently, megapascals (MPa). OneMPa equals approximately 9.9 atmospheres. Pressure isequivalent to a force per unit area (1 Pa = 1 N m–2) and toan energy per unit volume (1 Pa = 1 J m–3). A newton (N) =1 kg m s–1. Table 3.1 compares units of pressure.

If instead of pushing on the plunger we pull on it, a ten-sion, or negative hydrostatic pressure, develops in the waterto resist the pull. How hard must we pull on the plungerbefore the water molecules are torn away from each otherand the water column breaks? Breaking the water columnrequires sufficient energy to break the hydrogen bonds thatattract water molecules to one another.

Careful studies have demonstrated that water in smallcapillaries can resist tensions more negative than –30 MPa(the negative sign indicates tension, as opposed to com-pression). This value is only a fraction of the theoretical ten-sile strength of water computed on the basis of the strengthof hydrogen bonds. Nevertheless, it is quite substantial.

The presence of gas bubbles reduces the tensile strengthof a water column. For example, in the syringe shown inFigure 3.5, expansion of microscopic bubbles often inter-feres with the ability of the water to resist the pull exertedby the plunger. If a tiny gas bubble forms in a column ofwater under tension, the gas bubble may expand indefi-nitely, with the result that the tension in the liquid phasecollapses, a phenomenon known as cavitation. As we willsee in Chapter 4, cavitation can have a devastating effecton water transport through the xylem.

WATER TRANSPORT PROCESSESWhen water moves from the soil through the plant to theatmosphere, it travels through a widely variable medium(cell wall, cytoplasm, membrane, air spaces), and the mech-anisms of water transport also vary with the type ofmedium. For many years there has been much uncertainty

CapForce

Water Plunger

FIGURE 3.5 A sealed syringe can be used to create positiveand negative pressures in a fluid like water. Pushing on theplunger compresses the fluid, and a positive pressurebuilds up. If a small air bubble is trapped within thesyringe, it shrinks as the pressure increases. Pulling on theplunger causes the fluid to develop a tension, or negativepressure. Any air bubbles in the syringe will expand as thepressure is reduced.

TABLE 3.1Comparison of units of pressure

1 atmosphere = 14.7 pounds per square inch= 760 mm Hg (at sea level, 45° latitude)= 1.013 bar= 0.1013 Mpa= 1.013 × 105 Pa

A car tire is typically inflated to about 0.2 MPa.The water pressure in home plumbing is typically 0.2–0.3 MPa.The water pressure under 15 feet (5 m) of water is about

0.05 MPa.

about how water moves across plant membranes. Specifi-cally it was unclear whether water movement into plantcells was limited to the diffusion of water molecules acrossthe plasma membrane’s lipid bilayer or also involved dif-fusion through protein-lined pores (Figure 3.6).

Some studies indicated that diffusion directly across thelipid bilayer was not sufficient to account for observedrates of water movement across membranes, but the evi-dence in support of microscopic pores was not compelling.This uncertainty was put to rest with the recent discoveryof aquaporins (see Figure 3.6). Aquaporins are integralmembrane proteins that form water-selective channelsacross the membrane. Because water diffuses fasterthrough such channels than through a lipid bilayer, aqua-porins facilitate water movement into plant cells (Weig etal. 1997; Schäffner 1998; Tyerman et al. 1999). Note thatalthough the presence of aquaporins may alter the rate ofwater movement across the membrane, they do not changethe direction of transport or the driving force for watermovement. The mode of action of aquaporins is beingacitvely investigated (Tajkhorshid et al. 2002).

We will now consider the two major processes in watertransport: molecular diffusion and bulk flow.

Diffusion Is the Movement of Molecules byRandom Thermal AgitationWater molecules in a solution are not static; they are in con-tinuous motion, colliding with one another and exchang-ing kinetic energy. The molecules intermingle as a result of

their random thermal agitation. This random motion iscalled diffusion. As long as other forces are not acting onthe molecules, diffusion causes the net movement of mol-ecules from regions of high concentration to regions of lowconcentration—that is, down a concentration gradient(Figure 3.7).

In the 1880s the German scientist Adolf Fick discoveredthat the rate of diffusion is directly proportional to the con-centration gradient (∆cs/∆x)—that is, to the difference inconcentration of substance s (∆cs) between two points sep-arated by the distance ∆x. In symbols, we write this rela-tion as Fick’s first law:

(3.1)

The rate of transport, or the flux density (Js), is theamount of substance s crossing a unit area per unit time(e.g., Js may have units of moles per square meter per sec-ond [mol m–2 s–1]). The diffusion coefficient (Ds) is a pro-portionality constant that measures how easily substances moves through a particular medium. The diffusion coeffi-cient is a characteristic of the substance (larger moleculeshave smaller diffusion coefficients) and depends on themedium (diffusion in air is much faster than diffusion in aliquid, for example). The negative sign in the equation indi-cates that the flux moves down a concentration gradient.

Fick’s first law says that a substance will diffuse fasterwhen the concentration gradient becomes steeper (∆cs islarge) or when the diffusion coefficient is increased. Thisequation accounts only for movement in response to a con-centration gradient, and not for movement in response toother forces (e.g., pressure, electric fields, and so on).

Diffusion Is Rapid over Short Distances butExtremely Slow over Long DistancesFrom Fick’s first law, one can derive an expression for thetime it takes for a substance to diffuse a particular distance.If the initial conditions are such that all the solute mole-cules are concentrated at the starting position (Figure3.8A), then the concentration front moves away from thestarting position, as shown for a later time point in Figure3.8B. As the substance diffuses away from the startingpoint, the concentration gradient becomes less steep (∆csdecreases), and thus net movement becomes slower.

The average time needed for a particle to diffuse a dis-tance L is equal to L2/Ds, where Ds is the diffusion coeffi-cient, which depends on both the identity of the particleand the medium in which it is diffusing. Thus the averagetime required for a substance to diffuse a given distanceincreases in proportion to the square of that distance. Thediffusion coefficient for glucose in water is about 10–9 m2

s–1. Thus the average time required for a glucose moleculeto diffuse across a cell with a diameter of 50 µm is 2.5 s.However, the average time needed for the same glucosemolecule to diffuse a distance of 1 m in water is approxi-

J Dcxs ss= − ∆

∆

Water and Plant Cells 37

fpo

CYTOPLASM

OUTSIDE OF CELL

Water-selectivepore (aquaporin)

Water molecules

Membranebilayer

FIGURE 3.6 Water can cross plant membranes by diffusionof individual water molecules through the membranebilayer, as shown on the left, and by microscopic bulk flowof water molecules through a water-selective pore formedby integral membrane proteins such as aquaporins.

Chapter 3 38

0

Co

nce

ntr

atio

n

0

Co

nce

ntr

atio

n

(B)

Distance Dx Distance Dx

(A)

Time

DcsDcs

FIGURE 3.8 Graphical representation of the concentration gradient of a solute that isdiffusing according to Fick’s law. The solute molecules were initially located in theplane indicated on the x-axis. (A) The distribution of solute molecules shortly afterplacement at the plane of origin. Note how sharply the concentration drops off asthe distance, x, from the origin increases. (B) The solute distribution at a later timepoint. The average distance of the diffusing molecules from the origin has increased,and the slope of the gradient has flattened out. (After Nobel 1999.)

FIGURE 3.7 Thermal motion of molecules leads to diffusion—the gradual mixing ofmolecules and eventual dissipation of concentration differences. Initially, two mate-rials containing different molecules are brought into contact. The materials may begas, liquid, or solid. Diffusion is fastest in gases, slower in liquids, and slowest insolids. The initial separation of the molecules is depicted graphically in the upperpanels, and the corresponding concentration profiles are shown in the lower panelsas a function of position. With time, the mixing and randomization of the moleculesdiminishes net movement. At equilibrium the two types of molecules are randomly(evenly) distributed.

Initial Intermediate Equilibrium

Co

nce

ntr

atio

n

Position in container

Concentration profiles

mately 32 years. These values show that diffusion in solu-tions can be effective within cellular dimensions but is fartoo slow for mass transport over long distances. For addi-tional calculations on diffusion times, see Web Topic 3.2.

Pressure-Driven Bulk Flow Drives Long-DistanceWater TransportA second process by which water moves is known as bulkflow or mass flow. Bulk flow is the concerted movementof groups of molecules en masse, most often in response toa pressure gradient. Among many common examples ofbulk flow are water moving through a garden hose, a riverflowing, and rain falling.

If we consider bulk flow through a tube, the rate of vol-ume flow depends on the radius (r) of the tube, the viscos-ity (h) of the liquid, and the pressure gradient (∆Yp/∆x)that drives the flow. Jean-Léonard-Marie Poiseuille(1797–1869) was a French physician and physiologist, andthe relation just described is given by one form ofPoiseuille’s equation:

(3.2)

expressed in cubic meters per second (m3 s–1). This equa-tion tells us that pressure-driven bulk flow is very sensitiveto the radius of the tube. If the radius is doubled, the vol-ume flow rate increases by a factor of 16 (24).

Pressure-driven bulk flow of water is the predominantmechanism responsible for long-distance transport of waterin the xylem. It also accounts for much of the water flowthrough the soil and through the cell walls of plant tissues.In contrast to diffusion, pressure-driven bulk flow is inde-pendent of solute concentration gradients, as long as vis-cosity changes are negligible.

Osmosis Is Driven by a Water Potential GradientMembranes of plant cells are selectively permeable; thatis, they allow the movement of water and other smalluncharged substances across them more readily than themovement of larger solutes and charged substances (Stein1986).

Like molecular diffusion and pressure-driven bulk flow,osmosis occurs spontaneously in response to a drivingforce. In simple diffusion, substances move down a con-centration gradient; in pressure-driven bulk flow, sub-stances move down a pressure gradient; in osmosis, bothtypes of gradients influence transport (Finkelstein 1987).The direction and rate of water flow across a membrane aredetermined not solely by the concentration gradient of water orby the pressure gradient, but by the sum of these two drivingforces.

We will soon see how osmosis drives the movement ofwater across membranes. First, however, let’s discuss theconcept of a composite or total driving force, representingthe free-energy gradient of water.

The Chemical Potential of Water Represents theFree-Energy Status of WaterAll living things, including plants, require a continuousinput of free energy to maintain and repair their highlyorganized structures, as well as to grow and reproduce.Processes such as biochemical reactions, solute accumula-tion, and long-distance transport are all driven by an inputof free energy into the plant. (For a detailed discussion ofthe thermodynamic concept of free energy, see Chapter 2on the web site.)

The chemical potential of water is a quantitative expres-sion of the free energy associated with water. In thermo-dynamics, free energy represents the potential for per-forming work. Note that chemical potential is a relativequantity: It is expressed as the difference between thepotential of a substance in a given state and the potentialof the same substance in a standard state. The unit of chem-ical potential is energy per mole of substance (J mol–1).

For historical reasons, plant physiologists have mostoften used a related parameter called water potential,defined as the chemical potential of water divided by thepartial molal volume of water (the volume of 1 mol ofwater): 18 × 10–6 m3 mol–1. Water potential is a measure ofthe free energy of water per unit volume (J m–3). Theseunits are equivalent to pressure units such as the pascal,which is the common measurement unit for water poten-tial. Let’s look more closely at the important concept ofwater potential.

Three Major Factors Contribute to Cell WaterPotentialThe major factors influencing the water potential in plantsare concentration, pressure, and gravity. Water potential issymbolized by Yw (the Greek letter psi), and the waterpotential of solutions may be dissected into individualcomponents, usually written as the following sum:

(3.3)

The terms Ys, Yp, and Yg denote the effects of solutes, pres-sure, and gravity, respectively, on the free energy of water.(Alternative conventions for components of water poten-tial are discussed in Web Topic 3.3.) The reference stateused to define water potential is pure water at ambientpressure and temperature. Let’s consider each of the termson the right-hand side of Equation 3.3.

Solutes. The term Ys, called the solute potential or theosmotic potential, represents the effect of dissolved soluteson water potential. Solutes reduce the free energy of waterby diluting the water. This is primarily an entropy effect;that is, the mixing of solutes and water increases the dis-order of the system and thereby lowers free energy. Thismeans that the osmotic potential is independent of the spe-cific nature of the solute. For dilute solutions of nondisso-

Y Y Y Yw s p g= + +

Volume flow rate = xpp

hr4

8

∆∆Y

Water and Plant Cells 39

ciating substances, like sucrose, the osmotic potential maybe estimated by the van’t Hoff equation:

(3.4)

where R is the gas constant (8.32 J mol–1 K–1), T is theabsolute temperature (in degrees Kelvin, or K), and cs is thesolute concentration of the solution, expressed as osmolal-ity (moles of total dissolved solutes per liter of water [molL–1]). The minus sign indicates that dissolved solutesreduce the water potential of a solution relative to the ref-erence state of pure water.

Table 3.2 shows the values of RT at various temperaturesand the Ys values of solutions of different solute concen-trations. For ionic solutes that dissociate into two or moreparticles, cs must be multiplied by the number of dissoci-ated particles to account for the increased number of dis-solved particles.

Equation 3.4 is valid for “ideal” solutions at dilute con-centration. Real solutions frequently deviate from the ideal,especially at high concentrations—for example, greaterthan 0.1 mol L–1. In our treatment of water potential, wewill assume that we are dealing with ideal solutions (Fried-man 1986; Nobel 1999).

Pressure. The term Yp is the hydrostatic pressure of thesolution. Positive pressures raise the water potential; neg-ative pressures reduce it. Sometimes Yp is called pressurepotential. The positive hydrostatic pressure within cells isthe pressure referred to as turgor pressure. The value of Ypcan also be negative, as is the case in the xylem and in thewalls between cells, where a tension, or negative hydrostaticpressure, can develop. As we will see, negative pressuresoutside cells are very important in moving water long dis-tances through the plant.

Hydrostatic pressure is measured as the deviation fromambient pressure (for details, see Web Topic 3.5). Remem-ber that water in the reference state is at ambient pressure,so by this definition Yp = 0 MPa for water in the standardstate. Thus the value of Yp for pure water in an openbeaker is 0 MPa, even though its absolute pressure isapproximately 0.1 MPa (1 atmosphere).

Gravity. Gravity causes water to move downwardunless the force of gravity is opposed by an equal andopposite force. The term Yg depends on the height (h) ofthe water above the reference-state water, the density ofwater (rw), and the acceleration due to gravity (g). In sym-bols, we write the following:

(3.5)

where rwg has a value of 0.01 MPa m–1. Thus a vertical dis-tance of 10 m translates into a 0.1 MPa change in waterpotential.

When dealing with water transport at the cell level, thegravitational component (Yg) is generally omitted becauseit is negligible compared to the osmotic potential and thehydrostatic pressure. Thus, in these cases Equation 3.3 canbe simplified as follows:

(3.6)

In discussions of dry soils, seeds, and cell walls, one oftenfinds reference to another component of Yw, the matricpotential, which is discussed in Web Topic 3.4.

Water potential in the plant. Cell growth, photosyn-thesis, and crop productivity are all strongly influenced bywater potential and its components. Like the body tem-perature of humans, water potential is a good overall indi-cator of plant health. Plant scientists have thus expendedconsiderable effort in devising accurate and reliable meth-ods for evaluating the water status of plants. Some of theinstruments that have been used to measure Yw, Ys, andYp are described in Web Topic 3.5.

Water Enters the Cell along a Water PotentialGradientIn this section we will illustrate the osmotic behavior of plantcells with some numerical examples. First imagine an openbeaker full of pure water at 20°C (Figure 3.9A). Because thewater is open to the atmosphere, the hydrostatic pressure ofthe water is the same as atmospheric pressure (Yp = 0 MPa).There are no solutes in the water, so Ys = 0 MPa; thereforethe water potential is 0 MPa (Yw = Ys + Yp).

Y Y Yw s p= +

Yg w= r gh

Ys s= −RTc

Chapter 3 40

TABLE 3.2 Values of RT and osmotic potential of solutions at various temperatures

Osmotic potential (MPa) of solutionwith solute concentration in mol L–1 water

Temperature RTa Osmotic potential(°C) (L MPa mol–1) 0.01 0.10 1.00 of seawater (MPa)

0 2.271 −0.0227 −0.227 −2.27 −2.620 2.436 −0.0244 −0.244 −2.44 −2.825 2.478 −0.0248 −0.248 −2.48 −2.830 2.519 −0.0252 −0.252 −2.52 −2.9

aR = 0.0083143 L MPa mol–1 K–1.

Water and Plant Cells 41

FIGURE 3.9 Five examples illustrating the concept of water potential and its com-ponents. (A) Pure water. (B) A solution containing 0.1 M sucrose. (C) A flaccid cell(in air) is dropped in the 0.1 M sucrose solution. Because the starting water poten-tial of the cell is less than the water potential of the solution, the cell takes up water.After equilibration, the water potential of the cell rises to equal the water potentialof the solution, and the result is a cell with a positive turgor pressure. (D)Increasing the concentration of sucrose in the solution makes the cell lose water.The increased sucrose concentration lowers the solution water potential, drawswater out from the cell, and thereby reduces the cell’s turgor pressure. In this casethe protoplast is able to pull away from the cell wall (i.e, the cell plasmolyzes)because sucrose molecules are able to pass through the relatively large pores of thecell walls. In contrast, when a cell desiccates in air (e.g., the flaccid cell in panel C)plasmolysis does not occur because the water held by capillary forces in the cellwalls prevents air from infiltrating into any void between the plasma membraneand the cell wall. (E) Another way to make the cell lose water is to press it slowlybetween two plates. In this case, half of the cell water is removed, so cell osmoticpotential increases by a factor of 2.

(A) Pure water (B) Solution containing 0.1 M sucrose

(C) Flaccid cell dropped into sucrose solution

0.1 M Sucrose solution

(D) Concentration of sucrose increased

(E) Pressure applied to cell

Applied pressure squeezesout half the water, thus doubling s from –0.732 to –1.464 MPa

Yp = 0 MPaYs = 0 MPaYw = Yp + Ys

= 0 MPa

Pure water Yp = 0 MPaYs = –0.244 MPaYw = Yp + Ys

= 0 – 0.244 MPa= –0.244 MPa

0.1 M Sucrose solution

Yp = 0 MPaYs = –0.732 MPaYw = –0.732 MPa

Flaccid cell

Cell after equilibrium

Yw = –0.244 MPaYs = –0.732 MPaYp = Yw – Ys = 0.488 MPa

Yp = 0.488 MPaYs = –0.732 MPaYw = –0.244 MPa

Turgid cell

Yw = –0.732 MPaYs = –0.732 MPaYp = Yw – Ys = 0 MPa

Cell after equilibrium

Y

Yp = 0 MPaYs = –0.732 MPaYw = –0.732 MPa

0.3 M Sucrose solution

Yw = –0.244 MPaYs = –0.732 MPaYp = Yw – Ys = 0.488 MPa

Cell in initial state

Yw = –0.244 MPaYs = –1.464 MPaYp = Yw – Ys = 1.22 MPa

Cell in final state

Now imagine dissolving sucrose in the water to a con-centration of 0.1 M (Figure 3.9B). This addition lowers theosmotic potential (Ys) to –0.244 MPa (see Table 3.2) anddecreases the water potential (Yw) to –0.244 MPa.

Next consider a flaccid, or limp, plant cell (i.e., a cellwith no turgor pressure) that has a total internal solute con-centration of 0.3 M (Figure 3.9C). This solute concentrationgives an osmotic potential (Ys) of –0.732 MPa. Because thecell is flaccid, the internal pressure is the same as ambientpressure, so the hydrostatic pressure (Yp) is 0 MPa and thewater potential of the cell is –0.732 MPa.

What happens if this cell is placed in the beaker con-taining 0.1 M sucrose (see Figure 3.9C)? Because the waterpotential of the sucrose solution (Yw = –0.244 MPa; see Fig-ure 3.9B) is greater than the water potential of the cell (Yw= –0.732 MPa), water will move from the sucrose solutionto the cell (from high to low water potential).

Because plant cells are surrounded by relatively rigidcell walls, even a slight increase in cell volume causes alarge increase in the hydrostatic pressure within the cell.As water enters the cell, the cell wall is stretched by thecontents of the enlarging protoplast. The wall resists suchstretching by pushing back on the cell. This phenomenonis analogous to inflating a basketball with air, except thatair is compressible, whereas water is nearly incompressible.

As water moves into the cell, the hydrostatic pressure,or turgor pressure (Yp), of the cell increases. Consequently,the cell water potential (Yw) increases, and the differencebetween inside and outside water potentials (∆Yw) isreduced. Eventually, cell Yp increases enough to raise thecell Yw to the same value as the Yw of the sucrose solution.At this point, equilibrium is reached (∆Yw = 0 MPa), andnet water transport ceases.

Because the volume of the beaker is much larger thanthat of the cell, the tiny amount of water taken up by thecell does not significantly affect the solute concentration ofthe sucrose solution. Hence Ys, Yp, and Yw of the sucrosesolution are not altered. Therefore, at equilibrium, Yw(cell)= Yw(solution) = –0.244 MPa.

The exact calculation of cell Yp and Ys requires knowl-edge of the change in cell volume. However, if we assumethat the cell has a very rigid cell wall, then the increase incell volume will be small. Thus we can assume to a firstapproximation that Ys(cell) is unchanged during the equili-bration process and that Ys(solution) remains at –0.732 MPa.We can obtain cell hydrostatic pressure by rearrangingEquation 3.6 as follows: Yp = Yw – Ys = (–0.244) – (–0.732)= 0.488 MPa.

Water Can Also Leave the Cell in Response to aWater Potential GradientWater can also leave the cell by osmosis. If, in the previousexample, we remove our plant cell from the 0.1 M sucrosesolution and place it in a 0.3 M sucrose solution (Figure3.9D), Yw(solution) (–0.732 MPa) is more negative than

Yw(cell) (–0.244 MPa), and water will move from the turgidcell to the solution.

As water leaves the cell, the cell volume decreases. As thecell volume decreases, cell Yp and Yw decrease also untilYw(cell) = Yw(solution) = –0.732 MPa. From the water potentialequation (Equation 3.6) we can calculate that at equilibrium,Yp = 0 MPa. As before, we assume that the change in cellvolume is small, so we can ignore the change in Ys.

If we then slowly squeeze the turgid cell by pressing itbetween two plates (Figure 3.9E), we effectively raise thecell Yp, consequently raising the cell Yw and creating a∆Yw such that water now flows out of the cell. If we con-tinue squeezing until half the cell water is removed andthen hold the cell in this condition, the cell will reach a newequilibrium. As in the previous example, at equilibrium,∆Yw = 0 MPa, and the amount of water added to the exter-nal solution is so small that it can be ignored. The cell willthus return to the Yw value that it had before the squeez-ing procedure. However, the components of the cell Ywwill be quite different.

Because half of the water was squeezed out of the cellwhile the solutes remained inside the cell (the plasmamembrane is selectively permeable), the cell solution isconcentrated twofold, and thus Ys is lower (–0.732 × 2 =–1.464 MPa). Knowing the final values for Yw and Ys, wecan calculate the turgor pressure, using Equation 3.6, as Yp= Yw – Ys = (–0.244) – (–1.464) = 1.22 MPa. In our examplewe used an external force to change cell volume without achange in water potential. In nature, it is typically the waterpotential of the cell’s environment that changes, and thecell gains or loses water until its Yw matches that of its sur-roundings.

One point common to all these examples deservesemphasis: Water flow is a passive process. That is, water movesin response to physical forces, toward regions of low water poten-tial or low free energy. There are no metabolic “pumps” (reac-tions driven by ATP hydrolysis) that push water from oneplace to another. This rule is valid as long as water is theonly substance being transported. When solutes are trans-ported, however, as occurs for short distances across mem-branes (see Chapter 6) and for long distances in the phloem(see Chapter 10), then water transport may be coupled tosolute transport and this coupling may move water againsta water potential gradient.

For example, the transport of sugars, amino acids, orother small molecules by various membrane proteins can“drag” up to 260 water molecules across the membrane permolecule of solute transported (Loo et al. 1996). Such trans-port of water can occur even when the movement isagainst the usual water potential gradient (i.e., toward alarger water potential) because the loss of free energy bythe solute more than compensates for the gain of freeenergy by the water. The net change in free energy remainsnegative. In the phloem, the bulk flow of solutes and waterwithin sieve tubes occurs along gradients in hydrostatic

Chapter 3 42

(turgor) pressure rather than by osmosis. Thus, within thephloem, water can be transported from regions with lowerwater potentials (e.g., leaves) to regions with higher waterpotentials (e.g., roots). These situations notwithstanding, in thevast majority of cases water in plants moves from higher to lowerwater potentials.

Small Changes in Plant Cell Volume Cause LargeChanges in Turgor PressureCell walls provide plant cells with a substantial degree ofvolume homeostasis relative to the large changes in waterpotential that they experience as the everyday consequenceof the transpirational water losses associated with photo-synthesis (see Chapter 4). Because plant cells have fairlyrigid walls, a change in cell Yw is generally accompaniedby a large change in Yp, with relatively little change in cell(protoplast) volume.

This phenomenon is illustrated in plots of Yw, Yp, andYs as a function of relative cell volume. In the example ofa hypothetical cell shown in Figure 3.10, as Yw decreasesfrom 0 to about –2 MPa, the cell volume is reduced by only5%. Most of this decrease is due to a reduction in Yp (byabout 1.2 MPa); Ys decreases by about 0.3 MPa as a resultof water loss by the cell and consequent increased concen-tration of cell solutes. Contrast this with the volumechanges of a cell lacking a wall.

Measurements of cell water potential and cell volume(see Figure 3.10) can be used to quantify how cell wallsinfluence the water status of plant cells.

1. Turgor pressure (Yp > 0) exists only when cells arerelatively well hydrated. Turgor pressure in most cellsapproaches zero as the relative cell volume decreasesby 10 to 15%. However, for cells with very rigid cellwalls (e.g., mesophyll cells in the leaves of manypalm trees), the volume change associated with turgorloss can be much smaller, whereas in cells withextremely elastic walls, such as the water-storing cellsin the stems of many cacti, this volume change maybe substantially larger.

2. The Yp curve of Figure 3.10 provides a way to measurethe relative rigidity of the cell wall, symbolized by e(the Greek letter epsilon): e = ∆Yp/∆(relative volume). eis the slope of the Yp curve. e is not constant butdecreases as turgor pressure is lowered because nonlig-nified plant cell walls usually are rigid only when tur-gor pressure puts them under tension. Such cells actlike a basketball: The wall is stiff (has high e) when theball is inflated but becomes soft and collapsible (e = 0)when the ball loses pressure.

3. When e and Yp are low, changes in water potentialare dominated by changes in Ys (note how Yw andYs curves converge as the relative cell volumeapproaches 85%).

Water Transport Rates Depend on Driving Forceand Hydraulic ConductivitySo far, we have seen that water moves across a membranein response to a water potential gradient. The direction offlow is determined by the direction of the Yw gradient, andthe rate of water movement is proportional to the magni-tude of the driving gradient. However, for a cell that expe-riences a change in the water potential of its surroundings(e.g., see Figure 3.9), the movement of water across the cellmembrane will decrease with time as the internal andexternal water potentials converge (Figure 3.11). The rateapproaches zero in an exponential manner (see Dainty1976), with a half-time (half-times conveniently character-ize processes that change exponentially with time) givenby the following equation:

(3.7)

where V and A are, respectively, the volume and surface of

tA Lp

V1 2

0 693= ( )( )

−

.

e Ys

Water and Plant Cells 43

0.9 0.8

–3

–2

–1

0

1

2

1.0 0.95 0.85C

ell w

ater

po

ten

tial

(M

Pa)

Relative cell volume (DV/V)

Slope = e =DYp

DV/V

Zero turgor

Full turgor pressure

Yw = Ys + Yp

Ys

Yp

FIGURE 3.10 Relation between cell water potential (Yw)and its components (Yp and Ys), and relative cell volume(∆V/V). The plots show that turgor pressure (Yp) decreasessteeply with the initial 5% decrease in cell volume. In com-parison, osmotic potential (Ys) changes very little. As cellvolume decreases below 0.9 in this example, the situationreverses: Most of the change in water potential is due to adrop in cell Ys accompanied by relatively little change inturgor pressure. The slope of the curve that illustrates Ypversus volume relationship is a measure of the cell’s elasticmodulus (e) (a measurement of wall rigidity). Note that e isnot constant but decreases as the cell loses turgor. (AfterTyree and Jarvis 1982, based on a shoot of Sitka spruce.)

the cell, and Lp is the hydraulic conductivity of the cellmembrane. Hydraulic conductivity describes how readilywater can move across a membrane and has units of vol-ume of water per unit area of membrane per unit time perunit driving force (i.e., m3 m–2 s–1 MPa–1). For additionaldiscussion on hydraulic conductivity, see Web Topic 3.6.

A short half-time means fast equilibration. Thus, cellswith large surface-to-volume ratios, high membrane

hydraulic conductivity, and stiff cell walls (large e) willcome rapidly into equilibrium with their surroundings.Cell half-times typically range from 1 to 10 s, althoughsome are much shorter (Steudle 1989). These low half-timesmean that single cells come to water potential equilibriumwith their surroundings in less than 1 minute. For multi-cellular tissues, the half-times may be much larger.

The Water Potential Concept Helps Us Evaluatethe Water Status of a PlantThe concept of water potential has two principal uses: First,water potential governs transport across cell membranes,as we have described. Second, water potential is often usedas a measure of the water status of a plant. Because of tran-spirational water loss to the atmosphere, plants are seldomfully hydrated. They suffer from water deficits that lead toinhibition of plant growth and photosynthesis, as well asto other detrimental effects. Figure 3.12 lists some of thephysiological changes that plants experience as theybecome dry.

The process that is most affected by water deficit is cellgrowth. More severe water stress leads to inhibition of celldivision, inhibition of wall and protein synthesis, accumu-

Chapter 3 44Y

w (

MPa

)

Time

0

0

–0.2

Transport rate (Jv) slowsas Yw increases

D w =0.2 MPa

DYw =0.1 MPa

t1/2 = 0.693V(A)(Lp)(e –Ys)

(B)

Ψ

Yw = –0.2 MPaYw = 0 MPaDYw = 0.2 MPa

Initial Jv = Lp (DYw) = 10–6 m s–1 MPa–1

× 0.2 MPa = 0.2 × 10–6 m s–1

(A)

Water flow

FIGURE 3.11 The rate of water transport into a cell depends on thewater potential difference (∆Yw) and the hydraulic conductivity of thecell membranes (Lp). In this example, (A) the initial water potentialdifference is 0.2 MPa and Lp is 10–6 m s–1 MPa–1. These values give aninitial transport rate (Jv) of 0.2 × 10–6 m s–1. (B) As water is taken upby the cell, the water potential difference decreases with time, leadingto a slowing in the rate of water uptake. This effect follows an expo-nentially decaying time course with a half-time (t1/2

) that depends onthe following cell parameters: volume (V), surface area (A), Lp, volu-metric elastic modulus (e), and cell osmotic potential (Ys).

Abscisic acid accumulation

Physiological changesdue to dehydration:

Solute accumulation

Photosynthesis

Stomatal conductance

Protein synthesis

Wall synthesis

Cell expansion

Water potential (MPa)

Well-wateredplants

Pure water Plants undermild waterstress

Plants in arid,desert climates

–1–0 –2 –3 –4

FIGURE 3.12 Water potential of plantsunder various growing conditions,and sensitivity of various physiologi-cal processes to water potential. Theintensity of the bar color correspondsto the magnitude of the process. Forexample, cell expansion decreases aswater potential falls (becomes morenegative). Abscisic acid is a hormonethat induces stomatal closure duringwater stress (see Chapter 23). (AfterHsiao 1979.)

lation of solutes, closing of stomata, and inhibition of pho-tosynthesis. Water potential is one measure of howhydrated a plant is and thus provides a relative index ofthe water stress the plant is experiencing (see Chapter 25).

Figure 3.12 also shows representative values for Yw atvarious stages of water stress. In leaves of well-wateredplants, Yw ranges from –0.2 to about –1.0 MPa, but theleaves of plants in arid climates can have much lower val-ues, perhaps –2 to –5 MPa under extreme conditions.Because water transport is a passive process, plants cantake up water only when the plant Yw is less than the soilYw. As the soil becomes drier, the plant similarly becomesless hydrated (attains a lower Yw). If this were not the case,the soil would begin to extract water from the plant.

The Components of Water Potential Vary withGrowth Conditions and Location within the PlantJust as Yw values depend on the growing conditions andthe type of plant, so too, the values of Ys can vary consid-erably. Within cells of well-watered garden plants (exam-ples include lettuce, cucumber seedlings, and bean leaves),Ys may be as high as –0.5 MPa, although values of –0.8 to–1.2 MPa are more typical. The upper limit for cell Ys is setprobably by the minimum concentration of dissolved ions,metabolites, and proteins in the cytoplasm of living cells.

At the other extreme, plants under drought conditionssometimes attain a much lower Ys. For instance, waterstress typically leads to an accumulation of solutes in thecytoplasm and vacuole, thus allowing the plant to main-tain turgor pressure despite low water potentials.

Plant tissues that store high concentrations of sucrose orother sugars, such as sugar beet roots, sugarcane stems, orgrape berries, also attain low values of Ys. Values as low as–2.5 MPa are not unusual. Plants that grow in saline envi-ronments, called halophytes, typically have very low val-ues of Ys. A low Ys lowers cell Yw enough to extract waterfrom salt water, without allowing excessive levels of saltsto enter at the same time. Most crop plants cannot survivein seawater, which, because of the dissolved salts, has alower water potential than the plant tissues can attainwhile maintaining their functional competence.

Although Ys within cells may be quite negative, theapoplastic solution surrounding the cells—that is, in thecell walls and in the xylem—may contain only low con-centrations of solutes. Thus, Ys of this phase of the plant istypically much higher—for example, –0.1 to 0 MPa. Nega-tive water potentials in the xylem and cell walls are usuallydue to negative Yp. Values for Yp within cells of well-watered garden plants may range from 0.1 to perhaps 1MPa, depending on the value of Ys inside the cell.

A positive turgor pressure (Yp) is important for two prin-cipal reasons. First, growth of plant cells requires turgorpressure to stretch the cell walls. The loss of Yp under waterdeficits can explain in part why cell growth is so sensitive towater stress (see Chapter 25). The second reason positive

turgor is important is that turgor pressure increases themechanical rigidity of cells and tissues. This function of cellturgor pressure is particularly important for young, non-lignified tissues, which cannot support themselves mechan-ically without a high internal pressure. A plant wilts(becomes flaccid) when the turgor pressure inside the cellsof such tissues falls toward zero. Web Topic 3.7 discussesplasmolysis, the shrinking of the protoplast away from thecell wall, which occurs when cells in solution lose water.

Whereas the solution inside cells may have a positive andlarge Yp, the water outside the cell may have negative val-ues for Yp. In the xylem of rapidly transpiring plants, Ypis negative and may attain values of –1 MPa or lower. Themagnitude of Yp in the cell walls and xylem varies consid-erably, depending on the rate of transpiration and the heightof the plant. During the middle of the day, when transpira-tion is maximal, xylem Yp reaches its lowest, most negativevalues. At night, when transpiration is low and the plantrehydrates, it tends to increase.

SUMMARYWater is important in the life of plants because it makes upthe matrix and medium in which most biochemicalprocesses essential for life take place. The structure andproperties of water strongly influence the structure andproperties of proteins, membranes, nucleic acids, and othercell constituents.

In most land plants, water is continually lost to theatmosphere and taken up from the soil. The movement ofwater is driven by a reduction in free energy, and watermay move by diffusion, by bulk flow, or by a combinationof these fundamental transport mechanisms. Water diffusesbecause molecules are in constant thermal agitation, whichtends to even out concentration differences. Water movesby bulk flow in response to a pressure difference, wheneverthere is a suitable pathway for bulk movement of water.Osmosis, the movement of water across membranes,depends on a gradient in free energy of water across themembrane—a gradient commonly measured as a differ-ence in water potential.

Solute concentration and hydrostatic pressure are the twomajor factors that affect water potential, although when largevertical distances are involved, gravity is also important.These components of the water potential may be summed asfollows: Yw = Ys + Yp + Yg. Plant cells come into waterpotential equilibrium with their local environment by absorb-ing or losing water. Usually this change in cell volume resultsin a change in cell Yp, accompanied by minor changes in cellYs. The rate of water transport across a membrane dependson the water potential difference across the membrane andthe hydraulic conductivity of the membrane.

In addition to its importance in transport, water poten-tial is a useful measure of the water status of plants. As wewill see in Chapter 4, diffusion, bulk flow, and osmosis all

Water and Plant Cells 45

help move water from the soil through the plant to theatmosphere.

Web Material

Web Topics3.1 Calculating Capillary Rise

Quantification of capillary rise allows us to assessthe functional role of capillary rise in water move-ment of plants.

3.2 Calculating Half-Times of Diffusion

The assessment of the time needed for a mole-cule like glucose to diffuse across cells, tissues,and organs shows that diffusion has physiologi-cal significance only over short distances.

3.3 Alternative Conventions for Components ofWater Potential

Plant physiologists have developed several con-ventions to define water potential of plants. Acomparison of key definitions in some of theseconvention systems provides us with a betterunderstanding of the water relations literature.

3.4 The Matric Potential

A brief discussion of the concept of matric poten-tial, used to quantify the chemical potential ofwater in soils, seeds, and cell walls.

3.5 Measuring Water Potential

A detailed description of available methods tomeasure water potential in plant cells and tissues.

3.6 Understanding Hydraulic Conductivity

Hydraulic conductivity, a measurement of themembrane permeability to water, is one of thefactors determining the velocity of water move-ments in plants.

3.7 Wilting and Plasmolysis

Plasmolysis is a major structural change resultingfrom major water loss by osmosis.

Chapter References

Dainty, J. (1976) Water relations of plant cells. In Transport in Plants,Vol. 2, Part A: Cells (Encyclopedia of Plant Physiology, NewSeries, Vol. 2.), U. Lüttge and M. G. Pitman, eds., Springer, Berlin,pp. 12–35.

Finkelstein, A. (1987) Water Movement through Lipid Bilayers, Pores,and Plasma Membranes: Theory and Reality. Wiley, New York.

Friedman, M. H. (1986) Principles and Models of Biological Transport.Springer Verlag, Berlin.

Hsiao, T. C. (1979) Plant responses to water deficits, efficiency, anddrought resistance. Agricult. Meteorol. 14: 59–84.

Loo, D. D. F., Zeuthen, T., Chandy, G., and Wright, E. M. (1996)Cotransport of water by the Na+/glucose cotransporter. Proc.Natl. Acad. Sci. USA 93: 13367–13370.

Nobel, P. S. (1999) Physicochemical and Environmental Plant Physiology,2nd ed. Academic Press, San Diego, CA.

Schäffner, A. R. (1998) Aquaporin function, structure, and expres-sion: Are there more surprises to surface in water relations?Planta 204: 131–139.

Stein, W. D. (1986) Transport and Diffusion across Cell Membranes. Aca-demic Press, Orlando, FL.

Steudle, E. (1989) Water flow in plants and its coupling to otherprocesses: An overview. Methods Enzymol. 174: 183–225.

Tajkhorshid, E., Nollert, P., Jensen, M. Ø., Miercke, L. H. W., O’Con-nell, J., Stroud, R. M., and Schulten, K. (2002) Control of the selec-tivity of the aquaporin water channel family by global orienta-tion tuning. Science 296: 525–530.

Tyerman, S. D., Bohnert, H. J., Maurel, C., Steudle, E., and Smith, J.A. C. (1999) Plant aquaporins: Their molecular biology, bio-physics and significance for plant–water relations. J. Exp. Bot. 50:1055–1071.

Tyree, M. T., and Jarvis, P. G. (1982) Water in tissues and cells. InPhysiological Plant Ecology, Vol. 2: Water Relations and CarbonAssimilation (Encyclopedia of Plant Physiology, New Series, Vol.12B), O. L. Lange, P. S. Nobel, C. B. Osmond, and H. Ziegler, eds.,Springer, Berlin, pp. 35–77.

Weather and Our Food Supply (CAED Report 20). (1964) Center forAgricultural and Economic Development, Iowa State Universityof Science and Technology, Ames, IA.

Weig, A., Deswarte, C., and Chrispeels, M. J. (1997) The major intrin-sic protein family of Arabidopsis has 23 members that form threedistinct groups with functional aquaporins in each group. PlantPhysiol. 114: 1347–1357.

Whittaker R. H. (1970) Communities and Ecosystems. Macmillan, New York.

Chapter 3 46